ForecastBench Baseline - Zero-Shot - Market: leaderboard
ForecastBench Baseline leaderboard, where the ForecastBench team runs models out of the box, zero-shot prompt. Market component: questions from prediction markets and forecasting platforms; difficulty-adjusted Brier Index on 0-100 (higher is better; 50 equals always forecasting 0.5). Superforecaster and public medians, naive baselines and LLM crowds are excluded.
Metric: Market Brier Index (difficulty-adjusted, higher is better). Source: www.forecastbench.org. Saturation forecast: Around 2031. 81 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kimi K3 (Max) | 61.6 | #35 (Kimi K3) |
| 2 | GPT-4.5 (Preview) | 61.3 | #260 |
| 3 | Claude Opus 4.1 (20250805) | 61 | #139 |
| 4 | Grok Beta | 60.9 | #497 |
| 5 | Grok 4.20 0309 (Reasoning) | 60.5 | #120 (Grok 4.20 0309) |
| 6 | O4 Mini (2025-04-16) | 60.4 | #183 |
| 7 | Claude Sonnet 4.5 | 60.3 | #148 |
| 8 | Grok 4.20 Beta (0309) (Reasoning) | 60.2 | #198 (Grok 4.20 Beta (0309)) |
| 9 | Llama 3.3 70B Instruct | 60.1 | #541 |
| 10 | O3 (2025-04-16) | 60 | #125 |
| 11 | GLM-5 | 59.9 | #149 |
| 12 | Claude Haiku 4.5 (20251001) | 59.9 | #264 |
| 13 | Grok 4.3 | 59.8 | #204 |
| 14 | GPT-5.5 (2026-04-23) | 59.8 | #25 |
| 15 | Llama 3.1 405B Instruct | 59.7 | #470 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=forecastbench-baseline-zero-shot-market · How It Works · Data refreshed daily, snapshot 2026-11-05.